Sedimentation Rates and Metal Content of Sediments in a Venezuelan Coral Reef

opossumoozeMechanics

Feb 21, 2014 (3 years and 5 months ago)

122 views

Pergamon
PII: S0025-326X(98)00080-0
Marine Pollution Bulletin, Vol. 38, No. 1, pp. 16-24, 1999
© 1999 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
0025-326X/99 $ - - see front matter
Sedimentation Rates and Metal
Content of Sediments in a Venezuelan
Coral Reef
C. BASTIDAS*, D. BONE and E.M. GARCIA
Depto. de Biologla de Organismos, Universidad Sim6n Bolivar. Apdo. 89000, Caracas 1080-A, Venezuela
The sedimentation rate and metal content in trapped
sediments were determined at four localities at the
Parque Nacional Morrocoy and for a sampling period
of 1 year. The sedimentation rate was inversely related
to the distance from the main terrestrial sources of
sediment and ranged from 44+7 to 281+46
g m- 2d -n (mean+SE) for the farthest two localities
(11 km apart). February through May showed the
highest sedimentation rate considering all localities.
Except for Ca and Hg, metals in sediments also varied
inversely to the distance from terrestrial sources and
ranged as follows: AI 1.01-2.57, Fe 0.70-2.08, Ca
22.3-34.8 in %; and Zn 44.6-77.9, V 19.9-41.8, Pb
18.1-35.6, Cr 18.1-31.9, Cu 6.8-40.3, Hg 0.269-0.325
in lag g i based on dry weight. A common source of
metal pollution is suggested from the positive and
significant correlation found between the concentra-
tions of most metals. Only Hg was not correlated with
any other metal and its concentration was relatively
constant for all localities. Based on our results we
speculate that reef environments of P.N. Morrocoy are
being affected by long-term continuous load of
sediment moderately contaminated, brought about by
drainage of the neighboring rivers. © 1999 Elsevier
Science Ltd. All rights reserved
Keywords: sediments; sediment pollution metals; coral
reefs; coastal zone; Venezuela
Coastal reef areas are being heavily impacted by
increasing sediments from terrestrial origin, including
run-off and river inputs. The detrimental effects of
sediments on the structure of coral reef communities is
well documented. However, these effects on specific
coral species have been more variable depending on
the type and size of sediment, frequency of the load,
clearing ability, colony morphology and species resist-
ance (Bak and Elgershuizen, 1976; Dodge and Lang,
1983; Rice and Hunter, 1992; Stafford-Smith and
Ormond, 1992). Although being different factors, the
*Cor r espondi ng aut hor. Tel.: 617 47534334; Fax: 617 47725852.
E-mil: cbast i da@ai ms.gav.au.
turbidity and the sediment load or sedimentation (as
we used here) affect reef corals (Kendall et al., 1983;
Telesnicki and Goldberg, 1995).
Besides the impact that increased sedimentation
alone produces on coral reef communities (reviewed in
Rogers, 1990; van Katwijk et al., 1993; Reichelt and
Jones, 1994; Hunter and Evans, 1995) there could be
important contributions by contaminants associated
with these sediments, especially on fringing coral reefs.
The toxic substances associated with sewage (Pastorok
and Bilyard, 1985; Tomasick and Sander, 1987), drilling
and dredging activities (Kendall et al., 1983; Brown et
al., 1990) have been poorly characterized, and those
associated with terrestrial run-off have been largely
ignored (but see Cort6s and Risk, 1985).
Heavy metals, which are important pollutants in
coastal waters, represent one of the potential toxic
substances associated with run-off. In the water
column, they are known to be mainly associated with
sediments and to the suspended particulate organic
matter (Saouter et al., 1993). Incorporation of metals
into sediments can limit their bioavailability, but
several studies have linked sediment metal contamina-
tion to detrimental effects on ecosystems. For example,
this detrimental effect has been well established for
benthic invertebrates of soft bottom communities
(Schlekat et al., 1992, 1994; Burton and MacPherson,
1995). For most organisms, and especially for sessile
ones, the suspended sediments are an important source
of metals either by direct contact or when settled out
from the water column. Furthermore, the remobiliza-
tion and resuspension of sediments may return
contaminants to the water column even after external
sources have been eliminated (Schlekat et al., 1992).
Our study site comprises a marine park area (Parque
Nacional Morrocoy) on the west coast of Venezuela.
Near this park, three important rivers and other small
tributaries (carlos) discharge their waters. The Tocuyo
River, which drains at the north of the Park, has a total
length of 315 km and drain area of 17770 km 2. Bone et
al. (1993) studied its influence on the coral reef
community and the sedimentation rates at various
northern sites of the park. Also these authors reviewed
16
Marine Pollution Bulletin
the historical aspects of man's intrusion into the park
area. The Tocuyo River is supposed to have little influ-
ence on our study site. However, the rivers Aroa and
Yaracuy discharge at a minimum distance of 10 and
22 km, respectively, from our study site, and together
drain an area of approximately 4000 km 2 with serious
problems of erosion due to bad agricultural practices
and deforestation. Also a variety of industries
discharge their wastes into these rivers and an
important tourism development is expanding in the
coastal zone near the park.
Due to the lack of information in the southern area
of the park, we stated the following major objectives
for the present work: (1) to determine the sedimenta-
tion rates in four coral reef localities that differ in their
distance to the main sources of terrestrial inputs, and
(2) to determine the metal content in sediments
derived from terrestrial run-off and resuspension of
bottom sediments at each locality.
Materials and Methods
Four localities - - Punta Brava (PB), Paiclfis (PA),
Playuela (PL) and Bajo Caimfin (BC) - - were evalu-
ated within the Parque Nacional Morrocoy (PNM),
which has a total area of 320 km 2 and is located on the
west coast of Venezuela (between 10o52 ' N and 69o16 '
W, Fig. 1). The most distant sites - - Punta Brava (PB)
and Bajo Caiman (BC) - - are 11 km apart. The mean
annual rainfall for the period 1965-95 at various conti-
nental stations near the park ranged between 936 and
1283 mm, however the evaporation rates are usually
twice these values ( > 2000 mm) (data from the Direc-
tion of Hydrology and Meteorology, Ministry of
Environment and Natural Resources).
Three sediment traps were placed at each locality on
the sandy bottom at the end of the reef slope, 1.5 to
2m apart, between 6 and 9m depth, except in PL
where the traps were settled in the reef itself. The
sediment traps consisted of 6.8 cm diameter plastic jars
placed inside a 4" PVC tube which was pushed into the
sediment until it remained in a stable upright position.
The mouths of the plastic jars laid approximately 40 cm
above the sediment surface. This set-up resulted in a
height to width ratio of six which minimized the
capture of the sediment resuspended from the bottom
and maximized the collection efficiency (Gardner,
1980). Thus, the sediment trapped represents the
Cayo
$ombre~
~x
p~
r n
5Km
Fig. 1 Area of study located on the north west coast of Venezuela.
Sample sites marked with filled circles.
17
Volume 38/Number 1/January 1999
sediment that settles out from the suspended particu-
late matter of the water column (i.e. downward flux of
suspended particulate matter; see Tomasick and
Sander, 1985) and/or from the resuspension of bottom
sediments due to relatively high swell (as a conse-
quence of the trap height). The plastic jars were acid-
cleaned before settled in the field. Ten surveys were
made in a year at approximately monthly intervals
(Table 1).
Once retrieved, the sediment was dried until
constant weight at 60°C, homogenized and thawed
using a 250-1~m sieve to eliminate the macro-inverte-
brates captured in the traps. Then, the organic matter
content of the sediment was determined through
weight loss after muffle ignition at 550°C for 1 h.
The metal content (A1, Ca, Cr, Cu, Fe, Hg, Pb, V,
Zn) of the sediment was determined in five of the ten
samples (* in Table 1). For each locality, approximately
1 g dry weight of sediment was taken in duplicate for
metal analysis, with three lectures for each replicate.
The Hg was analyzed by cold vapor atomic absorp-
tion spectrophotometry technique (AAS, Perkin-Elmer
2380). The sample was digested with 3 ml of HNO3,
3 ml of H2SO4 and 3 ml of HF in a water bath for 5 h
at 58°C. The other metals were analyzed by ICP-EAS
using 10 ml HNO3 and 5 ml HC1 for the digestion of
these samples. After digestion, the samples were
filtered with Whatman 51 and distilled water added to
a final volume of 50 ml. Acids of analytic grade were
used for analysis, and for preparation of standards and
blanks.
The sedimentation rate was log transformed to
obtain normal distribution of the data (Shapiro-Wilks
test). Each factor (Locality and Sampling time) was
analyzed separately using an ANOVA due to the signi-
ficance of interaction term. The homogeneity of the
variance was checked through Barlett's test for each
factor.
The metal content of the collected sediments was
analyzed for the factors Locality and Sampling time
separately due to interactions between these two
factors (except for Ca and Hg). Most metals were
transformed as followed to fulfill the normality
assumption of ANOVA: logarithm was applied to Al,
Cr, Fe, Hg, Pb, V, Zn; square root was applied to Ca,
and 1/log to Cu. Multiple comparisons were made
using Fisher's least significant difference test. Cu, Cr,
V, and Zn showed heterogeneous variances when
grouped for locality and the multiple comparisons were
made using the Games-Howell method (Sokai and
Rohlf, 1995). Cu also showed heterogeneous variance
when grouped by sample time. The Ca and Cr content
in the sediments were not determined for the first
sampling period. The organic matter content of the
trapped sediment was analyzed with a two way
ANOVA without replication for the locality and
sample period.
Results
Sedimentation rates
The grand mean of the sedimentation rate found was
162+27gm 2d-1 (+SE, n = 100) for the four local-
ities and the entire sampling period. The sedimentation
rate was significantly different between the two local-
ities closest to the terrestrial inputs (PB and PA) and
the two localities further apart (PL and BC) (F = 16.41
dr= 3, p<0.001). The two first localities did not differ
in their sedimentation rates (281+46 and 247_+86
g m- 2 d- 1, n = 22 and 26, respectively). The other two
localities, PL and BC, did not differ between their rates
either, and showed sedimentation rates of 94 +29 and
44+7 g m -2 d -1, respectively (n = 26), after pooling
the sampling periods.
The sedimentation rate also varied significantly
between sampling periods (F = 10.07 dr= 9, p <0.001).
The largest sedimentation rate was obtained in
Feb-Mar (5) followed by May .(7) (Fig. 2). The
multiple comparisons showed that the sedimentation
rates from June through October (8-10) were signifi-
cantly lower than the rest of the sampling periods,
whereas the period of Feb-May (or 5-7) had the
highest sedimentation rate (Table 2). In this last
period, the gradient observed from PB to PL was lost
(Fig. 2). However, the trend between extreme localities
(PB and BC) was maintained. A great variability
TABLE 1
Sediment collection dates and number of days that the sediment traps remained in the field. Sedimentation rate as g m- 2 d J (mean + se, n = 3)
is also reported for each locality and sampling period. The asterisks (*) show the samples used for metal content analyses.
Label Date Days PB PA PL B
1 * = 0 30/09/94 to 21/10/94 21 304+74 82_+33 28_+7 31 -+0.5
2 * = 1 to 17/12/94 57 377+77 107_+53 23_+5 48+ 1
3 to 12/01/95 26 240 56 + 13 47
4 * = 2 to 03/02/95 22 781 139 _+ 7 64 + 7 73 + 24
5 to 17/03/95 42 517+85 817+518 413_+111 56+8
6 * = 3 to 21/04/95 35 228 _+ 22 486 _+ 319 20 -+ 8 97 -+ 29
7 to 25/05/95 34 317 _+ 138 568 _+ 3~)7 270 -+ 24 37 _+ 9
8 to 18/07/95 54 44 -+ 34 109 _+ 53 39 _+ 14 9 _+ 4
9 to 10/09/95 54 16 25 _+ 5 23 + 5 9 _+ 3
10 * = 4 to 26/10/95 47 79 _+ 42 22 + 4 22 _+ 9 22 _+ 0.3
Total mean 281 + 46 247 _+ 86 94 + 29 44 + 7
18
Marine Pollution Bulletin
1400-, -,-
12110
1o0o
~800
|
o-
¢-
£
200
1 2 3 4 5 6 7 8 9 10
Fig. 2 Sedimentation rates ( gm- Zd -=) for the localities in the
following order PB, PA, PL and BC throughout sampling
dates 1 to 10 (shown as in Table 1). The vertical bars
represent the standard deviation of the mean.
between replicates was also observed during this
period.
The data from Dec-Jan (3) has low reliability,
because this period corresponds to the maximum wind
speed, indeed the water velocity or swell is higher too
and only 5 out of 12 sediment traps remained upright.
However, the data from Jan-Feb (4) which was
reliable enough, suggest that we did not miss a great
sedimentation event in the time scale used here.
The organic matter content of the trapped sediments
showed a mean value of 12.6-1-3.00% ( ±SD, n = 20)
on a dry-weight basis, but did not vary through the
sampling period nor among localities (F=2.122,
p = 0.141, df= 4; F = 1.520, p = 0.260, dr= 3, respec-
tively).
Metal content of suspended sediment
The metal content of the sediment collected in the
traps is shown in Table 3. The concentration of metals
was significantly different among localities and
sampling periods (Table 4). High to low metal concen-
tration gradient from PB to BC was found for the
sediment metal content, except for the Ca and Hg
concentrations (Table 3, Fig. 3). The Ca showed its
lowest concentration in PB suggesting higher calcar-
eous influence at the other localities. The Hg content
in the sediments was relatively constant for all the
localities.
The metal content of sediment varied significantly
between sampling periods, except for Ca (Table 3, Fig.
3). In general, April and October 95 were periods of
TABLE 2
Sedimentation rate multiple comparisons between sampling periods (least significant difference for the 95%). The asterisk (*) denotes a
significant difference (t7 <0.05) between the two sampling periods. Blank cells mean no significant difference between the pair of sampling
periods.
10 9 8 7 6 5 4 3 2
l * * * * *
2 * * * *
3 * *
4 * * *
5 * * * *
6 * * *
7 * * *
8
9
10
19
Vol ume 38/Number 1/January 1999
TABLE 3
Metal content in sediment collected in traps at different localities of study. The concentration is shown as a percentage for AI, Ca and Fe, and
in ~g g ~ for the remaining metals, all on a dry-weight basis. Mean, sample size and standard error in this sequence.
Locality AI% Ca% Cu Cr Fe% Hg V Pb Zn
PB 2.57 22.303 40.28 31.95 2.08 0.269 41.82 35.57 77.92
10 8 10 8 10 9 10 10 10
0.29 2.34 16.65 3.26 0.31 0.061 5.05 3.42 12.84
PA
PL
BC
1.58 34.18 7.29 21.87 1.06 0.161 25.20 20.96 35.02
10 8 10 8 10 11 10 10 10
0.11 5.43 0.68 1.27 0.13 0.022 1.43 1.07 1.47
1.46 29.54 6.30 23.95 1.13 0.241 29.47 22.04 36.65
9 7 9 7 9 11 9 9 9
0.21 1.24 0.95 2.95 0.18 0.041 4.83 2.77 4.39
1.01 34.79 6.79 18.14 0.70 0.325 19.90 17.89 44.64
10 8 10 8 10 10 10 10 10
0.06 2.14 1.46 0.49 0.10 0.055 1.10 1.06 8.10
low metal content in the sediments and differed from
the rest of the sampling periods for most metals (Fig.
3). The interaction term of the analysis indicates that
for each metal the variation between sampling periods
depended on the locality. The variability between
sampling periods was more evident for PB, where most
metals occurred in their highest concentration. In this
locality, Cu and Hg metal content in the sediment was
different between the 0 and 4 sampling period (or
October of 94 and 95, respectively), indicating inter-
annual variability. A high association between metals
was observed as a result of the high and significant
correlation between them (Table 5). Only the Hg
concentration was not correlated with any other metal.
Also a negative association was found between most
metals and the Ca concentrations, with the exception
of Zn and Hg.
Discussion
The mean sedimentation rate of 162_+27 g m -2 d 1
(n = 100) found in the present study for the four local-
ities is slightly higher than the maximum value
reported by Bone (1980) for various localities at the
Parque Nacional Morrocoy. This author found a range
of 2.4 to 123.9 g m -2 d i that is also narrower than
the range found in the present study (9-800 g m -2
d-1). However, he did not evaluate the sites closest to
the rivers' drainage (PB and PA), that showed the
highest values, especially from February through May,
exceeding the range of 3 to 370 g m 2 d- 1 reported by
Pastorok and Bilyard (1985) for Caribbean natural
habitats when measured over extended time periods.
The values we found are similar to those of other
Caribbean coastal areas that have been exposed to
heavy terrestrial sedimentation, as in Cort6s and Risk
(1985).
The southern Caribbean climate is strongly influ-
enced by the southward migration of the subtropical
high pressure belt between Apr-May and November
(Hands et al., 1993). Our study area usually shows two
peaks of relatively high precipitation rates: May-Jun
and November. The peak of May corresponded
relatively well with the sedimentation rate found in this
study. However, the peak that should correspond to
Nov-Dec was not clearly detected and instead a
relatively high peak was found in February. Although it
TABLE 4
ANOVA results for the metal content for the two factors analyzed, locality and sampling period. The squared multiple r (r 2) indicates the data
adjustment to the general lineal model of the two factors without the interaction term.
Locality Sampling period
Metal N r 2 F-ratio p F-ratio p
AI 39 0.614 12.550 0.000 2.866 0.039
Ca 31 0.750 6.199 0.006 0.602 0.624 n.s.
Cr 31 0.603 7.609 0.001 4.556 0.012
Cu 39 0.673 13.197 0.000 5.891 0.001
Fe 39 0.501 5.957 0.002 3.310 0.023
Hg 39 0.801 6.197 0.004 10.841 0.000
Pb 39 {).577 9.730 0.000 3.207 0.026
V 39 0.749 12.236 0.000 13.990 0.000
Zn 39 0.539 6.361 0.002 4.055 0.009
20
Marine Pollution Bulletin
is possible that the corresponding Nov-Dec rainfall
was not high enough to produce a significant change in
the sedimentation rate, the sedimentation rate of
Oct'94-Jan'95 (1-4) was slightly higher than Jun-
Oct'95.
In Fig. 2 we can see that the sedimentation rates of
periods 4 and 5 were high and similar for PB, however,
25O
2OO
150
100
50
0
PB
V
PA PL BC
180
160
140
120
100
80
60
40
20
0
Pb
for PA and PL they were markedly different. This
could be due to the currents predominantly driven by
the wind (as it should be from Dec through Mar)
which could limit the river loads reaching the localities
of study, except in PB. Another possible explanation
(not mutually exclusive) is that the high sedimentation
rate found in February corresponds solely to the wind
Hg
1.8.
1.6 ~1
1.4
1.0
0.8 ¢.:!
0.6
O.4
m
0.2 m
i
PB PA PL BC PB PA PL BC
4OO
35O
3OO
25O
2OO
150
100
50
0
PB
Zn Cr Ca
+ m _+_+
PA PL BC
D 140
D 120
0 ° IO0
0 t ; ' O/ ; ; I
PB PA PL BC PB PA PL BC
12
10
8
6
4
2
0
!
Fe
14
12
10
8
+ +
! !
2
PB PA PL BC
AI
25O
10
i i 1o
13
: : : , 0
Cu
Oct-94
PB PA PL BC PB PA PL
Nov-Dec U Feb-95 ~ Apt-95
Fig. 3 Concentration of metals (mean for duplicates) for each
locality (PB, PA, PL, BC) and for the five sampling periods on
legend. Concentrations are given in ~tg g- t on a dry-weight
basis of < 250 pm fraction. Fe, Al and Ca in percentage.
Oct-95
BC
21
Volume 38/Number 1/January 1999
TABLE 5
Spearman correlation values for the concentration of metals in the sediment.
A1 Fe Zn Pb V Cr Cu Hg
Fe 0.8941"**
Zn 0.6187"** 0.6174"**
Pb 0.9257*** 0.8676*** 0.7006***
V 0.8785*** 0.8162"** 0.7775*** 0.9092***
Cr 0.8616"** 0.7499*** 0.5818"* 0.8745*** (I.8429"**
Cu 0.5960** 0.5724** 0.6151"** 0.6383*** 0.4776* 0.5172"*
Hg 0.0225 n.s. 0.1680 n.s. 0.1885 n.s. 0.1264 n.s. 0.0616 n.s. (/.0196 n.s.
Ca -0.6632*** -0.7108"** -0.3161 n.s. -0.5942** -0.5181"* -0.5328**
0.0645 n.s.
-0.4309* -0.1644 n.s.
*,o<0.05, **p<0.01, ***p <0.001.
speed, which is maximum for Feb and Mar in the study
area (average for these months between 1969 and
1989:12.5 versus 6.1 km/h for October).
The differences in sedimentation rate and metal
content of the sediment between Oct 94 and Oct 95
suggest interannual variability, which may be caused by
year-to-year variations in precipitation and wind.
A limitation of the sedimentation rates reported
here is the time elapsed between sampling periods
which could lead to high heterogeneity within a
locality, especially during periods of high sedimentation
rates. However, great variability of suspended particu-
late matter in coastal and/or estuarine environments
seems to be common (e.g. Benoit et al., 1994). Also,
the low sampling frequency could have affected the
metal concentration results, due to processes such as
metal bioaccumulation and/or biotransformation by the
infauna found in the traps.
Our results suggest the sediment that reaches our
study sites was as contaminated as other areas known
to be relatively contaminated by heavy metals (e.g.
Subramanian and Mohanachandran, 1990; Guzm~in
and Jim6nez, 1992; Moyano et al., 1993), although
these comparisons are difficult due to different
analytical procedures, geochemical composition of
region soils and, type and size of the sediments. Also,
processes such as element recycling by phytoplankton
(Hamilton-Taylor et al., 1984) or by other chemical
routes (Morse, 1994), could alter the composition of
deposited sediments, the commonest reported, resul-
ting in differences between suspended and deposited
sediments. However, in some cases there are no im-
portant differences between suspended and deposited
sediments (Benoit et al., 1994), as seemed to be the
case in the present study where we found metal
concentrations in the range of other studies in this area
(Table 6). The AI concentration of these studies was
low compared to our results, however, they are similar
for other study areas in Central America (Guzmfin and
Jim6nez, 1992).
In general, the sediments suspended in the water
column at PB tended to be more polluted with heavy
metals than the sediment in the other study sites. The
high concentration of metals such as AI, Fe and Zn,
together with the highest sedimentation rates, showed
a great terrigenous influence in PB. The reef organisms
of this locality should be more affected by the sedimen-
tation rate and the content of metals in these
sediments than in the other three localities. However,
the heterogeneity at other spatial scales (e.g. tens of
meters) within each locality remains unknown.
Low sediment rate seemed to be associated with low
metal content, which suggests an important contribu-
tion of the suspended particulate .matter from run-off
rather than from resuspension. The direct relationship
between suspended particulate matter and metal
content has been found in various studies (e.g. Benoit
et al., 1994; Cossa et al., 1994). The determination of
sediment grain size in our samples would have been
useful to determine if this trend is solely the conse-
quence of the association of the metal to different sizc
of particles (e.g. Subramanian and Mohanachandran,
1990; Benoit et al., 1994) or also to the increased
terrestrial run-off of contaminated waters.
TABLE 6
Metal content ranges in sediments from the Parque Nacional Morrocoy (Venezuela).
Hg (ppm) Fe (%) AI (%) Cu (ppm) Pb (ppm) Zn (ppm) Ni (ppm) Cd (ppm) Cr (ppm) V (ppm) Ref.
0.157 0.4-1.5 15-28 32-111 50-507 35-111 2.5-12.5 20-52 MARNR (1994) ~'
<0.002-1.416 0.035-0.088 1.0-3.0 1.7-5.0 8.4-15.0 2.1-3.5 <0.2 4.9-23.8 5.0-11.0 COF unp. data h
<0.002-1.416 <0.001-0.304 1.0-26.9 < 1.0-3.8 0.8-612 0.8-16.8 <0.2-1.2 4.9-86.5 COF unp. data'
0.161-0.325 0.70-2.08 1.01-2.57 6-40 18-36 33-78 18-32 20-42 Present study °
~Study sites spread over the park.
bOnly Yaracuy and Aroa river mouths.
CAll sites within and close to the park.
°Range from the locality averages.
22
Marine Pollution Bulletin
Fig. 4 Scanned LANDSAT image (from March 1978) showing the
study site during terrestrial sediment inputs reaching the
nearest sampling sites. The image is shown in the same
position as in Fig. 1 and the locality of Punta Brava (PB) is
shown for reference. The sediments coming from the north
correspond to the Tocuyo river, whereas those coming from
the south correspond to the Aroa and Yaracuy rivers.
The high correl at i on bet ween met al concent rat i on in
sedi ment s suggests t hey coul d have t he t errest ri al i nput
as a common source. However, it is also know t hat Fe
and AI coul d favour t he f or mat i on of colloids t hat bi nd
met al s or t hat t he presence of humi c subst ances (in t he
t errest ri al i nput ) coul d favor this t rend ( Decho and
Luoma, 1994; Wood et al., 1995).
Personal (qual i t at i ve) observat i ons i ndi cat e t hat t he
reef st ruct ure bet ween t hese localities also differs. The
reef at PB occurs at shal l ower waters, has fewer
number of species, and less live cover t han t he ot her
localities. However, we suggest that t hese charact er-
istics are not a consequence of a hi gher sedi ment at i on
rat e al one, but also due to a hi gher met al cont ent of
t he sedi ment s t hat reach PB. Besides, ot her toxic
subst ances associ at ed with run-off sedi ment s, such as
pesticides, sewage, or even salinity, coul d be addi t i onal
factors det er mi ni ng this trend. To t he mul t i pl e
pat hways in which sedi ment s may affect t he reef corals
and coral communi t i es, such as smot heri ng, bact eri al
infection, reduct i on of avai l abl e light and food capt ure,
and i ncr ement of energy expendi t ure for sedi ment
rej ect i on, t he toxic composi t i on and cont ent of this
sedi ment must be also consi dered.
Our results support t he i mpor t ance of t he t errest ri al
i nput s (run-off and river di scharges) in t he sedi ment a-
tion rat e of four coral reef localities within a mar i ne
pr ot ect ed area. Al so t hese t errest ri al i nput s seem
i mpor t ant in t hei r cont ri but i on to t he met al cont ent of
t he sedi ment set t l ed out from t he wat er col umn and
r esuspended from bot t om sedi ment s. A common
source of met al pol l ut i on at t he study site is also
suppor t ed by t he significant and positive correl at i on
found bet ween t he concent rat i on of most met al s.
Al t hough we do not provi de di rect evi dence that
such sedi ment s are ori gi nat ed from t errest ri al sources,
we can observe in t he LANDSAT i mages t hat t he
pl ume of sedi ment s brought over by t he nearby
sout hern rivers drai nage, do reach t he Par que Naci onal
Mor r ocoy area (Fig. 4). We suggest that t he occur-
rence of t hese cont i nuous events has been responsi bl e
over time, for t he progressi ve det er i or at i on of t he reef
envi ronment s of this park.
The authors wish to thank H. Guzm~in and R. Jaff6 for a critical
review of early versions of this manuscript. We also thank M.
Edreida for his help collecting the sediment traps. ICP measure-
ments were possible through the collaboration of O. Nufiez and the
Toxic Wastes Laboratory at U.S.B.
Bak, R. P. M. and Elgershuizen, H. B. W. (1976) Patterns of
oil-sediment rejection in corals. Mar. BioL 37, 105-113.
Benoit, G., Oktay-Marshall, S. D., Cantu, A., Hood, E. M., Coleman,
C. H., Corapcioglu, M. O. and Santschi, P. H. (1994) Partitioning
of Cu, Pb, Ag, Zn, Fe, AI and Mn between filter-retained particles,
colloids, and solution in six Texas estuaries. Mar. Chem. 45,
307-336.
Bone, D. (1980) Impacto de las actividades del hombre sobre los
arrecifes coralinos del Parque Nacionel Morrocoy. Tesis de
Licenciatura, Universidad Central de Venezuela, Caracas, 120 pp.
Bone, D., Losada, F. and Well, E. (1993) Origin of sedimentation
and its effect on the coral communities of a Venezuelan national
park. Ecotropicos 6, 10-21.
Burton, G. A. and MacPherson, C. (1995). Sediment toxicity testing
issues and methods. In Handbook of Ecotoxicology, eds D. J.
Hoffman, B. A. Rattner, G. A. Burton and J. Cairns, pp. 70-103.
Lewis Publishers, Boca Raton.
Brown, B. E., Le Tissier, M. D. A., Scoffin, T. P. and Tudhope, A.
W. (1990) Evaluation of the environmental impact of dredging on
intertidal coral reefs at Ko Phuket, Thailand, using ecological and
physiological parameters. Mar. Ecol. Prog. Ser. 65, 273-281.
Cortes, J. and Risk, M. J. (1985) A reef under siltation stress:
Cahuita, Costa Rica. Bull. Mar. Sci. 36, 339-356.
Decho, A. W. and Luoma, S. N. (1994) Humic and fulvic acids: sink
or source in the availability of metals to the marine bivalves
Macoma balthica and Potamocorbula amurensis? Mar. Ecol. Prog.
Ser. 108, 133-145.
Dodge, R. E. and Lang, J. C. (1983) Environmental correlates of
hermatypic coral (Montastrea annularis) growth on the East Fower
Gardens Bank, northwest Gulf of Mexico. Limnol. Oceanogr. 28,
228-240.
Gardner, W. D. (1980) Field assessment of sediment traps. Journal of
Marine Research 38, 41-52.
Guzm~in, H. M. and Jim6nez, C. E. (1992) Contamination of coral
reefs by heavy metals along the Caribbean coast of Central
America (Costa Rica and Panama). Mar. Pollut. Bull. 24, 554-561.
Hamilton-Taylor, J., Willis, M. and Reynolds, C. S. (1984) Deposi-
tional fluxes of metals and phytoplankton in Windermere as
measured by sediment traps. Limnol. Oceanogr. 29, 695-710.
Hands, M. R., French, J. R. and O'Neill, A. (1993) Reef stress at
Cahuita Point, Costa Rica: anthropogenically enhanced sediment
influx or natural geomorphic change? Journal of Coastal Research
9, 11-25.
Hunter, C. L, and Evans, C. W. (1995) Coral reefs in Kaneohe Bay,
Hawaii: two centuries of western influence and two decades of
data. Bull. Mar. Sci. 57(2), 501-515.
Kendall, J. J. Jr., Powell, E. N., Connor, S. J. and Bright, T. J. (1983)
The effects of drilling fluids (muds) and turbidity on the growth
and metabolic state of the coral Acropora cervicornis, with
23
Volume 38/Number 1/January 1999
comments on methods of normalization for coral data. Bull. Mar.
Sci. 33, 336-352.
Morse, J. W. (1994) Interactions of trace metals with authigenic
sulfide minerals: implications for their bioavailability. Mar. Chem.
46, 1-6.
MARNR (1994) Evaluaci6n de la calidad de las aguas del Parque
Nacional Morrocoy. (PT) Serie de Informes T6cnicos DGSCA/
IT/12, ISBN 98-04-1041, Caracas.
Moyano, M., Moresco, H., Blanco, J., Rosadilla, M. and Caballero,
A. (1993) Baseline studies of castall pollution by heavy metals, oil
and PAHs in Montevideo. Mar. Pollut. Bull. 26, 461-464.
Pastorok, R. A. and Bilyard, G. R. (1985) Effects of sewage pollution
on coral-reef communities. Mar. Ecol. Prog. Ser. 21, 175-189.
Reichelt, A. J. and Jones, G. B. (1994) Trace metals as tracers of
dredging activities in Cleveland Bay - - field and Laboratory
studies. Austr. J. Mar. Freshwater Res. 45, 1237-1257.
Rice, S. A. and Hunter, C. L. (1992) Effects of suspended sediment
and burial on scleractinian corals from west central Florida patch
reefs. Bull. Mar. Sci. 51, 429-442.
Rogers, C. S. (t990) Responses of coral reefs and reef organisms to
sedimentation. Mar. Ecol. Prog. Ser. 62, 185-202.
Saouter, E., Campbell, P. G. C., Ribeyre, F. and Boudou, A. (1993)
Use of partial extractions to study mercury partitioning on natural
sediment particles - - a cautionary note. Inter J. Eviron. AnaL
Chem. 54, 57-68.
Schlekat, C. E., McGee, B. L. and Reinharz, E. (1992) Testing
sediment toxicity in Chesapeake Bay with the amphipod Lepto-
cheirus plumulosus: an evaluation. Environ. Toxicol. Chem. 11,
225-236.
Schlekat, C. E., Megee, B. L., Boward, D. M., Reinharz, E.,
Velinsky, D. J. and Wade, T. L.. (1994) Tidal river sediments in
the Washington D.C. area. Biological effects associated with
sediment contamination. Estuaries 17(2), 334-344.
Sokal, R. R. and Rohlf, F. J. (1995) Biometry, 3rd ed. W. H.
Freeman and Company, New York.
Stafford-Smith, M. G. and Ormond, R. F. G. (1992) Sediment-
rejection mechanisms of 42 species of Australian scleractinan
corals. Aust. Z Mar. Freshwater Res. 43, 683-705.
Subramanian, V. and Mohanachandran, G. (1990) Heavy metals
distribution and enrichment in the sediments of southern east
coast of India. Mar. Pollut. Bull. 21, 324-330.
Telesnicki, G. J. and Goldberg, W. M. (1995) Effects of turbidity on
the photosynthesis and respiration of two south Florida reef coral
species. Bull. Mar. Sci. 57, 527-539.
Tomasick, T. and Sander, F. (1985) Effects of eutrophication on
reef-building corals. I. Growth rate of the reef-building coral
Montastrea annularis. Mar. Biol. 87, 143-155.
Tomasick, T. and Sander, F. (1987) Effects of eutrophication on
reef-building corals. II. Structure of scleractinian coral communi-
ties on fringing reefs, Barbados, West Indies. Mar. Biol. 94, 53-75.
van Katwijk, M. M., Meier, N. F., van Loon, R., van Hove, E. M.,
Giesen, W. B. J. T., van der Velde, G. and den Hartog, C. (1993)
Sabaki river sediment load and coral stress: correlation between
sediments and condition of the Malindi-Watamu reefs in Kenya
(Indian Ocean). Mar Biol. 117, 675-683.
Wood, T. M., Baptista, A. M., Kuwabara, J. S. and Flegal, A. R.
(1995) Diagnostic modeling of trace metal partitioning in south
San Francisco Bay. Limnol. Oceano~ 40, 345-358.
24